The work

You will assess the accuracy, naturalness, and cultural fit of Ukrainian text generated by a large language model. Day-to-day tasks may include evaluating translations and summaries, correcting dialogue or grammar, rewriting weak outputs, annotating grammatical, syntactic, and semantic features, and documenting recurring model errors for AI researchers.

What the screen is likely to test

Expect questions about specific Ukrainian-language projects and close analysis of examples involving grammar, syntax, stylistics, register, regional variation, and semantic fidelity. The screen may also test whether you can apply consistent evaluation criteria, distinguish an outright error from an acceptable variant, and explain findings clearly in English. Interest in AI and LLM-driven language technologies is preferred but not required.

Practical details

The role is designed for work in a distributed, remote team. The listing does not specify whether the workflow is asynchronous, the expected weekly commitment, or typical working hours, so candidates should be ready to state realistic availability and clarify scheduling requirements during the process.

Pay band: $40/hr